Download Pseudomonas aeruginosa B-band lipopolysaccharide genes wbpA

FEMS Microbiology Letters 189 (2000) 135^141
www.fems-microbiology.org
Pseudomonas aeruginosa B-band lipopolysaccharide genes wbpA and
wbpI and their Escherichia coli homologues wecC and wecB are not
functionally interchangeable
Lori L. Burrows
a
a;b
, Kathryn E. Pigeon c , Joseph S. Lam
c;
*
Received 21 March 2000; received in revised form 19 May 2000 ; accepted 1 June 2000
Abstract
The O antigen unit of Pseudomonas aeruginosa serotype O5 is a complex trisaccharide containing 2-acetamido-3-acetiminido-2,3-dideoxyL-D-mannuronic acid, 2-acetimido-3-acetimido-2,3-dideoxy-L-D-mannuronic acid, and 2-acetimido-2,6-deoxy-L-D-galactosamine. Specific
knockout mutations in the putative UDP-D-N-acetylglucosamine (UDP-D-GlcNAc) epimerase gene, wbpI, or the putative UDP-D-Nacetylmannosamine dehydrogenase gene, wbpA, resulted in strains that no longer produced B-band lipopolysaccharide, confirming the
essential roles of these genes in B-band O antigen synthesis. Despite approximately 50% similarity of wbpI and wbpA to the Escherichia coli
genes wecB (rffE) and wecC (rffD) involved in enterobacterial common antigen synthesis, cross-complementation experiments were not
successful. These results imply that the P. aeruginosa UDP-D-GlcNAc precursor may be di-N-acetylated prior to further modification,
preventing the E. coli enzymes from recognizing it as a substrate. ß 2000 Federation of European Microbiological Societies. Published by
Elsevier Science B.V. All rights reserved.
Keywords : WbpA; Dehydrogenase; WbpI; Epimerase; O antigen; Lipopolysaccharide ; Enterobacterial common antigen; Pseudomonas aeruginosa
1. Introduction
Pseudomonas aeruginosa co-produces two forms of lipopolysaccharide (LPS), the common antigen, A-band LPS,
and the serotype-speci¢c antigen, B-band LPS. In previous
work [1], we identi¢ed 16 genes thought to be involved in
synthesis of the trisaccharide B-band O unit of serotype
O5. We proposed a putative pathway [1] for the biosynthesis of the three monosaccharides of the O unit, 2-acetamido-3-acetiminido-2,3-dideoxy-L-D-mannosaminuronic
acid (Man(2NAc3N)A), 2-acetimido-3-acetimido-2,3-dideoxy-L-D-mannosaminuronic acid (Man(2NAc3NAc)A),
and 2-acetimido-2,6-dideoxy-L-D-galactosamine (Fuc2NAc) [2]. We proposed that the two mannosaminuronic
acid residues of this trisaccharide were generated through
a common series of biochemical steps, followed by further
alteration of the ¢rst residue to the 3-acetiminido form
through an unknown mechanism [1]. This latter step is
* Corresponding author. Tel. : +1 (519) 824-4120 ext. 3823 ;
Fax: +1 (519) 837-1802; E-mail: [email protected]
now thought to be catalyzed by the product of wbpG [3].
The order of the preceding steps in this series of reactions,
however, is not unequivocal. For example, WbpA, a putative UDP-mannose dehydrogenase, was suggested to use
UDP-D-Man2NAc3NAc as a substrate [1] but could conceivably act at other points in the pathway to ultimately
convert UDP-D-Man2NAc (or UDP-D-Man2NAc3N) to
UDP-D-Man(2NAc3NAc)A. For that reason, it is di¤cult
to begin to assess the activity of this and other enzymes of
the pathway, since the identity of appropriate substrates is
unclear. In addition, the unusual substrates and intermediates for these enzymes are not commercially available,
which complicates analysis of their activities.
To understand the synthesis of UDP-D-Man(2NAc3NAc)A and its acetiminido derivative, we set out to create
knockout mutants de¢cient in each step of the pathway.
These mutants would facilitate the examination of the
roles played by WbpI, a protein with homology to
UDP-N-acetyl-D-glucosamine (UDP-D-GlcNAc)-2-epimerases, and WbpA, a putative UDP-D-Man2NAc3NAc dehydrogenase in the synthesis of this rare mannosaminuronic acid and its derivative. WbpA has 54% similarity with
0378-1097 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 8 - 1 0 9 7 ( 0 0 ) 0 0 2 6 3 - 9
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
Center for Infection and Biomaterials Research, Toronto General Hospital, Toronto, Ont., Canada
b
Department of Surgery, University of Toronto, Toronto, Ont., Canada
c
Department of Microbiology, University of Guelph, Guelph, Ont., Canada
136
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
2. Materials and methods
2.1. Bacterial strains, plasmids and culture conditions
The bacterial strains and plasmids used in this study are
listed in Table 2. P. aeruginosa and E. coli were grown in
Luria broth (LB; Gibco BRL) or on agar plates at 37³C
unless otherwise stated. Pseudomonas isolation agar (PIA;
Difco) was used to select transconjugants following mating
experiments. The antibiotic concentrations used were: ampicillin at 100 Wg ml31 for E. coli and carbenicillin (Cb) at
300 Wg ml31 for P. aeruginosa, gentamicin (Gm) at 15 Wg
ml31 for E. coli and 300 Wg ml31 for P. aeruginosa, tetracycline at 15 Wg ml31 for E. coli and 90 Wg ml31 for
P. aeruginosa, and kanamycin at 30 Wg ml31 for E. coli.
2.2. DNA methods
Plasmid DNA was isolated by the alkaline lysis method
of Birnboim and Doly [7] or the QIAprep spin mini-prep
kit (Qiagen). P. aeruginosa chromosomal DNA was isolated with DNAzol (Gibco BRL) using the manufacturer's
directions. Restriction and modi¢cation enzymes were
purchased from Gibco BRL, Boehringer Mannheim, and
Pharmacia and used according to the suppliers' speci¢cations. To transform E. coli and P. aeruginosa with plasmids, electrocompetent cells were prepared according to
Binotto et al. [8] and Burrows et al. [1], respectively, followed by electroporation using a Gene Pulser (Bio-Rad).
Homology searches of the National Center for Biotechnology DNA and protein database were conducted using the
BLAST (Basic Local Alignment Search Tool) server analysis programs [9].
2.3. LPS isolation and analysis
LPS was prepared as described by Hitchcock and
Brown [10] from overnight broth cultures. The LPS preparations were separated on SDS^PAGE gels and LPS pro¢les detected by silver staining [11]. Alternatively, LPS was
transferred to Biotrace nitrocellulose (Gelman, Rexdale,
Ont., Canada) and visualized by immunoblotting as described previously [1]. The LPS blots were incubated
with hybridoma culture supernatants containing one of
the monoclonal antibodies N1F10 (speci¢c for P. aeruginosa A-band LPS), MF15-4 (speci¢c for P. aeruginosa
O5 B-band LPS) or 898 (speci¢c for E. coli ECA; this
antibody was a generous gift of Paul Rick).
2.4. Generation of wbpI and wbpA insertional mutants
A strategy involving insertional mutation with gentamicin resistance cassette (GmR ) gene replacement was used
to produce chromosomal knockout mutants [12]. The gene
of interest (wbpI or wbpA) was cloned into the suicide
Table 1
Exopolysaccharide structures
a
Mannosaminuronic acid residues are indicated in bold.
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
WbpO, a UDP-D-N-acetylgalactosamine (GalNAc) dehydrogenase from P. aeruginosa serotype O6. The Salmonella
typhi gene wcdA (vipA), encoding the UDP-D-GalNAc dehydrogenase involved in the biosynthesis of the Vi capsular antigen, complemented O antigen synthesis in wbpO
knockout mutants [4], con¢rming that wbpO encodes a
UDP-D-GalNAc dehydrogenase.
WbpA and WbpI share approximately 50% amino acid
similarity to the Escherichia coli enzymes WecC (R¡D)
and WecB (R¡E), respectively, that are involved in the
synthesis of enterobacterial common antigen (ECA) (Table
1) [5]. ECA is a mannosaminuronic-acid-containing exopolysaccharide that can be attached to core lipid A in
Enterobacteriaceae. We have previously used an ECA-biosynthetic gene (wecA, encoding the UDP-D-GlcNAc glycosyltransferase responsible for initiating assembly of the
ECA polymer) to demonstrate, by cross-complementation,
the function and substrate speci¢city of the P. aeruginosa
O5 initiating glycosyltransferase, WbpL [6]. Here, we use a
similar strategy to determine whether wbpA and wbpI of
P. aeruginosa were functionally as well as genetically homologous to wecC and wecB of E. coli.
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
vector pEX100T [13]. pEX100T contains the Bacillus subtilis sacB gene, which when expressed in a Gram-negative
bacterium makes it sensitive to killing by sucrose. Streaking cells on medium containing both Gm and sucrose permits the growth only of recombinants that have undergone
a double crossover event. Knockout constructs were mobilized from E. coli SM10 to P. aeruginosa by the method
of Simon et al. [14].
2.5. PCR ampli¢cation and cloning of wecB and wecC
facilitate directional cloning. Primer sequences were: wecB
5P-TTGGGCGTCCAATGCTTCAGGCTCG-3P (wecb1,
upstream primer) and 5P-CATCTCTGCTAACCATTCTGCCATC-3P (wecb2, downstream primer). For wecC,
the upstream primer (wecc1) was 5P-TGGTAGGCATGCATAAGCAGCGAAT-3P and the downstream primer
(wecc2) was 5P-GCCACATCAGGATCCCCGCGTAGGT-3P.
The ampli¢cation of wecB and wecC from E. coli K-12
AB1133 [5] chromosomal DNA was performed using a
PowerBlock I thermocycler (Ericomp) and a high-¢delity
DNA polymerase, PwoI, (Boehringer Mannheim) according to the manufacturer's speci¢cations. The products obtained via PCR ampli¢cation were puri¢ed using the
Boehringer Mannheim High-Pure PCR Puri¢cation kit,
digested with the appropriate restriction enzymes, and ligated (using T4 DNA ligase) to pUCP27 that was previ-
Table 2
Bacterial strains and plasmids
Strain or plasmid
P. aeruginosa
PAO1
KEP-A3
KEP-I9
E. coli
JM109
SM10
AB1133
21546
21566
21685
Plasmids
pACYC184
pUCP20, pUCP21
pUCP26, pUCP27
pEX100T
pUCGM
pFV161-26a, b, pFV161-20
pFV180-26, pFV180-20
pFV161TG
pFV180TG
pWECC-27
pWECC-21
pWECC-184
pWECB-27
pWECB-21
pCA62
Genotype or relevant characteristics
Reference/
source
Serotype O5; A+ B+
Serotype O5, wbpA : :Gm, A+ B3
Serotype O5, wbpI: :Gm, A+ B3
[20]
This study
This study
recA1 supE44 endA1 hsdR17 gyrA96 relA1 thi vlac-proAB FP; (traD36, proAB‡ , lacIq , lacZvM15)
thi-1 thr leu tonA lacY supE recA RP4-2-Tc: :Mu; KmR
K-12; thr-1 leuB6 v(gpt-proA)66 hisG4 argE3 thi-1 rfbD1 lacY ara-14 galK2 xyl-5 mtl-1 mgl-51 rpsL31
kdgK51 supE44
AB1133 wecC (r¡D): :Tn10
AB1133 wecB: :Tn10Pa , wecC (r¡D): :Tn10
21566+pCA62 (r¡DE)
[21]
[14]
[5]
4.2-kb low-copy-number cloning vector containing the p15A origin of replication, CmRb , TetR
3.9-kb pUC18-derived broad-host-range cloning vectors with multiple cloning sites in opposite
orientations; ApR , CbR
4.9-kb pUC18-derived broad-host-range cloning vectors with multiple cloning sites in opposite
orientations; TcR
5.8-kb gene replacement vector, oriT‡ , sacB‡ ; ApR
Source of GmR cassette; ApR , GmR
2.7-kb XhoI-BamHI fragment in pUCP26 or pUCP20; contains wbpA oriented downstream of (a) or
opposite (b) the vector-encoded lac promoter
2.8-kb XbaI-PstI fragment inserted into pUCP26 or pUCP20; contains wbpI expressed from the
vector-encoded lac promoter
2.7-kb XhoI-BamHI fragment in pEX100T, with 0.9-kb GmR cassette from pUCGM inserted into
unique HindIII site within wbpA and oriented to permit expression of wbpB^E from the aacC1
promoter
2.8-kb XbaI-PstI fragment in pEX100T; with 0.9-kb GmR cassette from pUCGm inserted into unique
XhoI site within wbpI and oriented to permit expression of wbpJ^L from aacC1 promoter
pUCP27 containing wecC ampli¢ed by PCR from E. coli using primers wecc1 and wecc2 (see
Section 2)
same insert as above, cloned into pUCP21
V1.5-kb SphI-SmaI wecC fragment from pWECC-27 cloned into pACYC184 (SphI-SalI)
pUCP27 containing wecB ampli¢ed by PCR from E. coli using primers wecb1 and wecb2 (see
Section 2)
same insert as above, cloned into pUCP21
contains wecB and wecC
C. Whit¢eld
[18]
[5]
[5]
[5]
[18]
[13]
[22]
This work
This work
This work
This work
This work
This work
This work
This work
This work
[5]
a
Marolda and Valvano [23] showed by PCR that in 21566, wecB contains a small insert, hypothesized to be a fragment of an excised Tn10, in addition
to an insertion in the 5P region of wecC.
b
Ap, ampicillin ; Cb, carbenicillin ; Cm, chloramphenicol ; Gm, gentamicin ; Km, kanamycin ; Tc, tetracycline.
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
Oligonucleotide primers for ampli¢cation of the wecB
and wecC genes were designed using GeneRunner (Hastings Software, Inc.). Primers were synthesized at the University of Guelph Molecular Supercenter using a PerkinElmer 394 DNA synthesizer. PCR of wecB and wecC was
performed using primers incorporating restriction sites to
137
138
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
Fig. 1. Location of wbpA and wbpI (black arrows) within the P. aeruginosa serotype O5 B-band O antigen gene cluster (not to scale). The
sites of insertion of the GmR cassettes are indicated by open arrowheads.
3. Results and discussion
3.1. Generation and analysis of gene replacement mutants
of wbpA and wbpI
Many of the O antigens of P. aeruginosa contain complex and unusual amino sugars rarely seen elsewhere in
nature [2]. The biosynthetic pathways leading to the formation of these unusual monosaccharides are likely to be
similarly unique. Using previously described techniques,
we created mutants of P. aeruginosa PAO1 (serotype O5)
that contained non-polar gentamicin resistance cassettes
inserted into either wbpA or wbpI (Fig. 1).
Examination by silver-stained SDS^PAGE and Western
immunoblots of the LPS isolated from each mutant
showed that neither mutant produced B-band O antigen
(Fig. 2). Synthesis of A-band O antigen was not a¡ected.
Complementation of each mutant with the cognate gene
restored B-band O antigen synthesis to levels comparable
to the wild-type (Fig. 2). The cloned wbpA gene was able
to complement the wbpA : :Gm mutant regardless of orientation with respect to the vector-encoded promoter,
demonstrating the presence of a functional promoter
upstream of wbpA (Fig. 2, pFV161-26a and pFV16126b). These experiments showed that both WbpA and
WbpI are essential for serotype O5 B-band O antigen synthesis.
3.2. Cross-complementation experiments with E. coli ECA
mutants
The Staphylococcus aureus type 5 and type 8 capsules
Table 3
Amino acid homologies of WbpA and WbpI with their counterparts in
E. coli and S. aureus
Proteina
WbpA
WecC
Cap5O
WbpI
WecB
Cap5P
WecC
Cap5O
WbpI
WecB
Cap5P
Cap5G
32 (51)
33 (54)
n/a
n/a
n/a
n/a
45 (68)
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
33 (48)
31 (50)
33 (56)
50 (68)
nd
nd
n/a, not applicable; nd, not determined.
a
Homologies are given as % identity, with % similarity in brackets.
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
ously digested with the corresponding enzymes. For complementation of the E. coli mutants 21546 and 21566 (TcR
due to Tn10 insertion), a wbpA-containing insert and the
PCR products were cloned into the ApR vectors pUCP20
and pUCP21, respectively. For dual complementation of
the E. coli mutant 21566 with both wecB and wecC, the
wecC-containing insert was cloned into the TcR gene of
pACYC184 to form a chloramphenicol-resistant construct.
Similarly, the complementation of 21566 with wbpI and
wecC was tested using a wbpI-containing insert cloned
into pUCP20 to investigate the functional relatedness of
WecB and WbpI.
are chemically and structurally similar to the P. aeruginosa
serotype O5 B-band O antigen (Table 1), containing L-DManNAcA attached to D-Fuc2NAc, a 2,6-dideoxy derivative of D-GalNAc. Not surprisingly, the biosynthetic clusters for these polysaccharides share many homologous
genes [1,15]. We showed recently that S. aureus cap8D is
functionally homologous to the highly conserved P. aeruginosa gene wbpM [16]. In the case of E. coli, as shown in
Table 1, ECA also contains the amino sugar L-D-ManNAcA attached to L-D-GlcNAc. Therefore, all three sugar
polymers contain singly or doubly acetylated mannosaminuronic acid residues. The homologies of WbpA and
WbpI with their homologues in S. aureus and E. coli are
shown in Table 3.
Kiser and Lee [17] showed previously that the S. aureus
capsule genes cap5O and cap5P (but not the cap5P homologue cap5G) were able to complement E. coli wecC and
wecB mutants, respectively. However, they did not perform complementation experiments with the E. coli genes
in the S. aureus background. We performed cross-complementation experiments to examine the functional homology of wbpA and wbpI to their E. coli homologues wecC
and wecB. The cloned E. coli wecC gene was not able to
complement the P. aeruginosa wbpA: :Gm mutant (Fig. 2),
although it could complement its cognate E. coli mutant,
21546 (wecC: :Tn10; Fig. 3). Similarly, the cloned E. coli
wecB gene was unable to complement the P. aeruginosa
wbpI: :Gm mutant (Fig. 2), but complemented the E. coli
wecB mutant (Fig. 3).
In corresponding experiments, we showed that wbpA
could complement the P. aeruginosa wbpA: :Gm mutant
(Fig. 2) but not the E. coli wecC mutant (Fig. 3). Similarly,
wbpI could complement the wbpI: :Gm mutant (Fig. 2) but
not the wecB mutant (Fig. 3). The latter result is consistent
with the ¢ndings of Kiser and Lee [17], who showed that
cap5P, but not its homologue cap5G, could complement a
wecB mutant. Protein sequence analysis of WbpI showed
that it is slightly more similar to Cap5G than Cap5P (56%
similarity vs. 50% similarity).
Kiser and Lee [17] concluded that Cap5P was the UDPD-GlcNAc-2-epimerase responsible for synthesis of the DMan2NAcA residue of the capsule unit and speculated
that Cap5G could be involved in FucNAc synthesis. The
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
139
S. aureus capsule contains both L-FucNAc and D-FucNAc,
each of which must be synthesized via a separate pathway
from di¡erent precursors. Therefore, Cap5G may be involved in one of those two pathways. In P. aeruginosa, the
synthesis of UDP-D-FucNAc is thought to require wbpK,
wbpM and possibly wbpB [1,16] ,none of which has signi¢cant homology to wecB or cap5G.
In PAO1, wbpI is the only candidate UDP-D-GlcNAc-2epimerase gene, and is therefore strongly implicated in
UDP-D-Man(2NAc3NAc)A biosynthesis. Protein cluster
analysis using the SYSTERS database (http://www.
dkfz-heidelberg.de/tbi/services/cluster/systersform) shows
that WbpI, WecB, Cap5P and Cap5G all fall into a com-
mon group (cluster N423) of putative nucleotide sugar C2
isomerases. Inspection of amino acid sequence alignments
shows regions of homology throughout the proteins (not
shown). However, without any information regarding the
tertiary structure of these enzymes, it is di¤cult to make
any kind of observations with respect to function based on
the homology or lack thereof in particular regions of primary amino acid sequence.
The inability of wecB to complement a wbpI mutation
and vice versa could occur for a number of reasons. For
example, it is possible that the proteins are not expressed
in the heterologous background, that the encoded enzymes
cannot interact with heterologous biosynthetic machinery
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
Fig. 2. Mutation and complementation of wbpA and wbpI in P. aeruginosa O5. A: A wbpA knockout mutation abrogates B-band O antigen synthesis
while A-band LPS synthesis is not a¡ected. Note that more A-band LPS is produced in the absence of B-band LPS, and that the insert of pFV161
complements the mutation regardless of orientation, showing a functional promoter is present upstream of wbpA. Complementation with wbpA, but not
with wecC, restores B-band LPS synthesis. B: Loss of wbpI has a similar e¡ect on A- and B-band LPS expression, and can be complemented by wbpI,
but not wecB.
140
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
in a non-native background, or that their substrates are
su¤ciently di¡erent as to prevent complementation. We
have previously demonstrated that E. coli proteins, including WecA, could be expressed and proven to be functional
in the P. aeruginosa background [6]. In addition, all of the
cloned proteins used in these studies were expressed from
the vector-encoded E. coli lac promoter, which has previously been shown to function in P. aeruginosa [18]. Kiser
and Lee [17] showed that a cloned copy of wecB could
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
Fig. 3. Complementation of ECA biosynthesis by E. coli and P. aeruginosa genes. A: Complementation of ECA expression in an E. coli
wecC mutant with wecC but not wbpA. Note that the modulation of
ECA chain lengths is perturbed when wecC is provided on a high-copynumber plasmid. B: Complementation of ECA expression in an E. coli
wecB mutant. Since the transposon insertion in wecB is polar on wecC,
both genes need to be provided in trans to complement the mutation.
pCA62 contains both genes on a single insert [5], while each gene is
provided on separate vectors in the adjacent lane. Note that the perturbation of ECA chain length is apparent only when the genes are provided separately on vectors of disparate copy number (pACYC184 has
a lower copy number than pUCP21). ECA biosynthesis is not restored
by providing wecC and wbpI in trans.
complement the cap5P mutation in S. aureus, a Grampositive bacterium, implying that WecB was correctly expressed in that heterologous background. Finally, the integrity of the constructs used for complementation was
clearly demonstrated by their ability to complement their
cognate mutants. Taken together, these results suggest
that poor expression of the relevant proteins is unlikely
to explain the lack of complementation.
Instead, it is more likely that the substrate for WbpI is
not UDP-D-GlcNAc, as it is for WecB and Cap5P, but
possibly UDP-D-Glc2NAc3NAc. The presence of the additional N-acetyl group at C3 (the product of WbpD and
WbpE activities [3]) could prevent WecB and Cap5P from
utilizing this substrate, while requirement for this speci¢c
substitution would preclude WbpI from recognizing a
mono-N-acetylated UDP-D-GlcNAc substrate. Biochemical analysis of the nucleotide sugar precursors produced
by the wbpI mutant is required to test this hypothesis.
Supporting evidence for the ability of P. aeruginosa to
synthesize this particular di-N-acetylated nucleotide sugar
is the presence of D-Glc(2NAc3NAc)A in the O antigen of
serotype O1 (Table 1).
Similarly, the inability of wbpA to complement a wecC
mutant and vice versa is more likely to be due to the
di¡erences in their substrate speci¢cities, rather than to
an inability of the enzyme to be correctly expressed in a
non-native background. While the dehydrogenation step
catalyzed by WbpA probably occurs after the 2-epimerization catalyzed by WbpI, based on the E. coli and
S. aureus models, biochemical con¢rmation is required.
The O antigen of P. aeruginosa serotype O1 (Table 1)
contains D-Glc(2NAc3NAc)A, implying dehydrogenation
can occur without 2-epimerization. However, protein cluster analysis of WbpA using the SYSTERS database shows
that both WbpA and WecC fall into cluster O743, subfamily 2, which includes UDP-D-ManNAc dehydrogenases
and GDP-D-mannose dehydrogenases (not shown). The
remaining three subfamilies in this cluster contain only
UDP-D-glucose dehydrogenases. These results suggest
that the substrate of WbpA has a mannose, rather than
a glucose, con¢guration.
Interestingly, Bordetella pertussis, which produces a lipooligosaccharide containing D-Man(2NAc3NAc)A, does
not possess a wbpA homologue [19]. In that organism,
the dehydrogenase function is assigned to WlbA [19],
which has low similarity (39% over 351 aa) with WbpB,
postulated to encode an oxidoreductase in P. aeruginosa.
SYSTERS analysis of WlbA clearly clusters it with oxidoreductases (not shown), suggesting that another protein
encoded by B. pertussis may be responsible for its UDP-DMan2NAc3NAc dehydrogenase activity.
In summary, we have shown that wbpA and wbpI are
essential for biosynthesis of the serotype O5 B-band O
unit and that they are not functionally homologous to
their E. coli counterparts. Using these results, we can
now design logical experiments to test the biochemical
L.L. Burrows et al. / FEMS Microbiology Letters 189 (2000) 135^141
functions of these novel enzymes with an improved understanding of their most likely substrates and intermediates.
[10]
Acknowledgements
[11]
We thank P. Rick and R.Y.C. Lo for strains 21546,
21566, and for the anti-ECA monoclonal antibody 898.
This work was funded by Medical Research Council of
Canada Grant MT-14687 to J.S.L. L.L.B. was the recipient of a Canadian Cystic Fibrosis Foundation Postdoctoral Fellowship.
[12]
[13]
[14]
References
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSIBLAST : a new generation of protein database search programs. Nucleic Acids Res. 25, 3389^3402.
Hitchcock, P.J. and Brown, T.M. (1983) Morphological heterogeneity among Salmonella lipopolysaccharide chemotypes in silver-stained
polyacrylamide gels. J. Bacteriol. 154, 269^277.
Dubray, G. and Bezard, G. (1982) A highly sensitive periodic acidsilver stain for 1,2-diol groups of glycoproteins and polysaccharides
in polyacrylamide gels. Anal. Biochem. 119, 325^329.
de Kievit, T.R., Dasgupta, T., Schweizer, H. and Lam, J.S. (1995)
Molecular cloning and characterization of the rfc gene of Pseudomonas aeruginosa (serotype O5). Mol. Microbiol. 16, 565^574.
Schweizer, H.P. and Hoang, T.T. (1995) An improved system for
gene replacement and xylE fusion analysis in Pseudomonas aeruginosa. Gene 158, 15^22.
Simon, R., Priefer, U. and Pu«hler, A. (1983) A broad-host-range
mobilization system for in vivo genetic engineering: transposon mutagenesis in gram-negative bacteria. Bio/Technology 1, 784^791.
Sau, S., Bhasin, N., Wann, E.R., Lee, J.C., Foster, T.J. and Lee, C.Y.
(1997) The Staphylococcus aureus allelic genetic loci for serotype 5
and 8 capsule expression contain the type-speci¢c genes £anked by
common genes. Microbiology 143, 2395^2405.
Burrows, L.L., Urbanic, R.V. and Lam, J.S. (2000) Functional conservation of the polysaccharide biosynthetic protein WbpM and its
homologues in Pseudomonas aeruginosa and other medically signi¢cant bacteria. Infect. Immun. 68, 931^936.
Kiser, K.B. and Lee, J.C. (1998) Staphylococcus aureus cap5O and
cap5P genes functionally complement mutations a¡ecting enterobacterial common-antigen biosynthesis in Escherichia coli. J. Bacteriol.
180, 403^406.
West, S.E.H., Schweitzer, H.P., Dall, C., Sample, A.K. and RunyenJanecky, L.J. (1994) Construction of improved Escherichia-Pseudomonas shuttle vectors derived from pUC18/19 and the sequence of
the region required for their replication in Pseudomonas aeruginosa.
Gene 128, 81^86.
Allen, A. and Maskell, D. (1996) The identi¢cation, cloning and
mutagenesis of a genetic locus required for lipopolysaccharide biosynthesis in Bordetella pertussis. Mol. Microbiol. 19, 37^52.
Hancock, R.E. and Carey, A.M. (1979) Outer membrane of Pseudomonas aeruginosa: heat- and 2-mercaptoethanol-modi¢able proteins.
J. Bacteriol. 140, 902^910.
Yanisch-Perron, C., Vieira, J. and Messing, J. (1985) Improved M13
phage cloning vectors and host strains: nucleotide sequences of the
M13mp18 and pUC19 vectors. Gene 33, 103^119.
Schweizer, H.D. (1993) Small broad-host-range gentamycin resistance
gene cassettes for site-speci¢c insertion and deletion mutagenesis.
Biotechniques 15, 831^834.
Marolda, C.L. and Valvano, M.A. (1995) Genetic analysis of the
dTDP-rhamnose biosynthesis region of the Escherichia coli VW187
(O7:K1) rfb gene cluster : identi¢cation of functional homologs of
rfbB and rfbA in the r¡ cluster and correct location of the r¡E
gene. J. Bacteriol. 177, 5539^5546.
FEMSLE 9490 26-7-00
Downloaded from http://femsle.oxfordjournals.org/ by guest on October 25, 2016
[1] Burrows, L.L., Charter, D.F. and Lam, J.S. (1996) Molecular characterization of the Pseudomonas aeruginosa serotype O5 (PAO1) Bband lipopolysaccharide gene cluster. Mol. Microbiol. 22, 481^495.
[2] Knirel, Yu.A., Vinogradov, E.V., Kocharova, N.A., Paramonov,
N.A., Kochetkov, N.K., Dmitriev, B.A., Stanislavsky, E.S. and Lanyi, B. (1988) The structure of O-speci¢c polysaccharides and serological classi¢cation of Pseudomonas aeruginosa (a review). Acta Microbiol. Hung. 35, 3^24.
[3] Rocchetta, H.L., Burrows, L.L. and Lam, J.S. (1999) Genetics of Oantigen biosynthesis in Pseudomonas aeruginosa. Microbiol. Mol.
Biol. Rev. 63, 523^553.
[4] Belanger, M., Burrows, L.L. and Lam, J.S. (1999) Functional analysis of genes responsible for the synthesis of the B-band O antigen of
Pseudomonas aeruginosa serotype O6 lipopolysaccharide. Microbiology 145, 3505^3521.
[5] Meier-Dieter, U., Starman, R., Barr, K., Mayer, H. and Rick, P.D.
(1990) Biosynthesis of enterobacterial common antigen in Escherichia
coli. Biochemical characterization of Tn10 insertion mutants defective
in enterobacterial common antigen synthesis. J. Biol. Chem. 265,
13490^13497.
[6] Rocchetta, H.L., Burrows, L.L., Pacan, J.C. and Lam, J.S. (1998)
Three rhamnosyltransferases responsible for assembly of the A-band
D-rhamnan polysaccharide in Pseudomonas aeruginosa: a fourth
transferase, WbpL, is required for the initiation of both A-band
and B-band lipopolysaccharide synthesis. Mol. Microbiol. 28, 1103^
1119.
[7] Birnboim, H.C. and Doly, J. (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res.
7, 1513^1523.
[8] Binotto, J., MacLachlan, P.R. and Sanderson, K.E. (1991) Electrotransformation in Salmonella typhimurium LT2. Can. J. Microbiol.
37, 474^477.
[9] Altschul, S.F., Madden, T.L., Scha¡er, A.A., Zhang, J., Zhang, Z.,
141